I think that maybe what you’re trying to say is that a) only a few counties in a few states will decide the election; the rest are already foregone conclusions. And b) the undecided / independent (“UI”) vote that will decide those few counties is too volatile to predict this far out.
If I understand you correctly I agree with both those propositions.
But I also think that in addition to the few genuinely up-for-grabs areas there are a bunch of areas where voter turnout matters far more than what the UI voter does on election day. Turnout matters. Which is why the Rs are trying everything they can think of to suppress voting in general, and D voting in particular.
As much as the pollsters try, they have a very hard time predicting who will bother to vote. With the error bars on turnout among the UIs being especially large.
If I’ve misunderstood you and put words in your mouth I’m sorry. Please try again with more context.
I don’t say that, but Trump historically outperforms polls. Do reputable pollsters now give Biden a couple extra points to compensate? No, only a disreputable pollster would do that. After their 2016 miss, the good ones sought objective ways to avoid a Democratic ticket skew, but the skew again occurred in 2020. Most likely, shy Trumpers remain a reality.
Independents do think that Trump is a bad man. They even think he is a worse man than Biden. You can see this in the personal approval polling, where – unlike with the horse race question – Trump is behind.
But independents tend to value economic performance more highly and reject claims, someone like me makes, concerning presidents having little to no short-term influence over inflation rates and the business cycle. Thus – neck and neck in the horse race, with indictments having zero effect.
This is literally true, but state level polling is less accurate.
Also, these early polls are of registered voters. Pollsters will switch to likely voter models, that tend to favor the Republicans, next year.
Joe and Kamala are in trouble. Hiding this does no good.
A recent poll showed that 43% of respondents said that they would definitely not vote for Biden. The same poll said that 53% of respondents would definitely not vote for Trump. Another 11% said they probably wouldn’t vote for him. That’s 64% that definitely or probably won’t vote for Trump. If this is true, then Trump’s ceiling is 47% if he gets literally every other possible voter including the 11% that said they probably won’t. How does he win a general election with these numbers? Where does he get the independent and moderate voters that the GOP must get? The base is passionate, but the base alone can’t win you anything but a blood red district.
He wouldn’t, even though he can lose the popular election and still win (which he did in 2016) that’s too much of a disparity. In 2016, he got 46% of the popular vote to Hillary’s 48%, only a 2% difference there, and he only narrowly won the electoral vote. In 2020 he got 47%, and Biden got 51%, and Biden narrowly won the electoral vote. So I think that shows you how little wiggle room there is if you intend to win the election while still losing the popular vote.
But that assumes that the polls are accurate, and that they’ll stay steady by election time, and I doubt either of those things are true. (Of course, that leaves open the possibility that the polls are more in his favor than reality, and that opinion of him will decline between now and then, so I’m not pushing doom and gloom here.)
By those numbers, 57% might vote for Biden and 47% might vote for Trump. If 14% of the might-Biden voters in 2-3 swing states change their minds, stay home or vote third-party, while all the might-Trump voters vote Trump, Trump wins.
In 2016 there weren’t 53% of voters saying that they definitely would not vote for Trump this far from the election. This is not analogous to the circumstances in 2016.
I’m not trying to blow smoke up anyone’s butt here, and I’m not making any predictions (I retired from that after the 2016 election), but I also know that the GOP is paying for bullshit polls, just like they did before the “red wave” that wasn’t. Makes me wonder why we continue to look at polls from places that told us over and over leading up to Nov 2022 that the red tsunami was coming. Every poll from any outlet that did that should be forever ignored.
Agreed. But if we learned anything from 2016 it’s the perils of complacency.
Of course polls can be inaccurate – they undercounted Trump voters in 2016 and 2020, and overestimated the red tsunami in 2022 – but they’re not utter fantasy. They mostly show that almost half of voters would consider voting for Trump again, and “almost half” can be plenty.
Only if you count the 11% that said they probably won’t vote for him as “considering voting for him”. I guess technically, but I don’t see him getting all 11% to swing from probably won’t to will. So it’s not really close to half.
To me it’s depressing that even today so many people would be willing to vote for a massive criminal who actively conspired to harm the country and democracy itself, but at least it’s pretty unlikely he’ll ever sit in the White House again.
It’s lots simpler than that. It’s this and only this:
The worst imaginable R is better than the best imaginable D. Therefore I vote trump since he’s the R. I have no need to evaluate trump (or anyone else) beyond confirming there’s an (R) after his name.
@airbeck: The thing every poll skips, even the ones aimed at “likely voters”, is turnout.
In your first post on this sub-sub-sub topic this afternoon you said
I’d amend that to more like this
A recent poll showed that 43% of respondents said that they would definitely not vote for Biden if they bother to vote at all. The same poll said that 53% of respondents would definitely not vote for Trump if they bother to vote at all. Another 11% said they probably wouldn’t vote for him if they bother to vote at all. That’s 64% that definitely or probably won’t vote for Trump if they bother to vote at all.
The critical advantage trump has is enthusiasm. His voters are passionate and will turn up on election day. The critical Achilles heel for Biden is that many of his voters are not passionate. And may not show up on election day.
Preponderance of turnout will decide the election, not preponderance of preference. And given the EC advantages, the gerrymandering, the voter suppression, and the enthusiasm difference, it is easy for 70% of the Never-Bidens to vote while only 55% of the Never-trumps do. And if so trump wins.
Most significantly, it does not really matter what the national polling numbers say. It matters what they say in the handful of counties in competitive states that will decide this election. The national numbers only matter as they serve as sort of a monkey-see-monkey-do bellwether for the folks in the electorally significant counties to key off of. But they’ll only key off that national number or national momentum or national news so much. The error bars on their adherence to national numbers are large.
Change the right 10,000 votes in a country of 350M people and you change the outcome. And lots more of that changing is not in specific undecided voters choosing D vs. choosing R. It’s in choosing which 10,000 people show up to be counted.
Is it really those handful of counties that matter, or are they just the bellwether for the competitive states? I don’t think it’s whether or not the suburbs of Atlanta, Phoenix, Philadelphia, and Milwaukee go red or blue. It’s the overall turnout in those states.
For sure the statewide popular vote total determines which team’s delegates go to Washington for the real vote in the EC. As you say.
I would argue that some counties by virtue of size and purple-ness and the extra presence or absence of corrupt voting and vote-counting practices will have an outsized influence on the range of plausible statewide-totals for the state. Some areas have pretty predictable turnout numbers year after year. Others less so.
It’s my thesis that those relative wildcards will be the ones that really matter within the handful of states whose outcome really matters.
I’m not aware of any 2023 week when the rolling Trump-Biden polling average moved more than a point or two, or showed Biden ahead when you correct by the 3-4 point GOP electoral college advantage. The Carroll verdict did nothing, just as the four indictments did nothing. This fits my knowing, in real life, Trump voters, and Biden voters, and not knowing any undecided voters.
Now, my saying there are no undecided voters may be another version of polling results denial, since polls do show eleven percent or so undecided. But given Trump’s history of outperforming polls, it’s plausible that many of those telling pollsters they are undecided are actually shy Trumpers.
Can some event come along that changes this? Sure. But Biden and Trump have been neck and neck for so long that it is going to have to be one heck of an event.
My guess is the “shy Trump” voters don’t exist to any large degree anymore. Back in 2016, 2018, and maybe even 2020, sure. But after 1/6/2021 and the reaction to it by Trump, I think that’s done with. They’re now almost all non-shy Trump voters. I suspect the undecided voters mostly come in two types.
One group is the old fashioned traditional “Eisenhower Republican.” The sort of person who’s a fan of the National Review. They were lifelong Republicans, even voting Trump in 2016. Then they voted for Biden in 2020. Now that Trump is out of office and they aren’t confronted daily with Trump policy, but are reminded daily of Democratic policy, they aren’t sure which way to vote.
The other is the far left voter, who tends to vote D when there’s a Republican incumbent but votes Green or Independent when there’s a Democratic incumbent.
I suppose there’s also some low information voters who genuinely don’t pay attention to politics, but I assume polls screen out that kind of voter.